Background: Venous thromboembolism (VTE) is a major cause of morbidity and mortality in the United States. The increased risk of VTE among medically ill older adults is firmly established. Guidelines strongly promote the use of pharmacologic VTE prophylaxis in this population, yet studies have suggested sub-optimal prophylaxis rates. Our study aimed to describe current provider practices with regard to VTE prophylaxis and identify intrinsic and extrinsic risk factors for, as well as associated outcomes of, prescribing noncompliance in a cohort of hospitalized older adults.
Methods: A retrospective study was conducted of hospitalized adults aged ≥75 years, admitted to the medicine service of a large academic tertiary center from May 1, 2014 to June 30, 2015. Process and outcome variables were extracted from electronic health records (EHR). The primary outcome was noncompliance, defined as the absence of an order for VTE prophylaxis for the duration of hospitalization or an interruption of prophylaxis exceeding 24 hours. Secondary measures included in-hospital mortality, length of stay (LOS), and 30-day readmissions. Chi-squared tests were used to evaluate associations between compliance and relevant variables and outcomes.
Results: Of 3,751 patients, 58.8% were female, 62.9% white, with a mean age of 84.7. An order for prophylaxis was recorded in 97.6% of patients; 11.0% were found to be noncompliant. Pharmacologic agents with or without mechanical prophylaxis were prescribed for 83.3% of patients, while mechanical prophylaxis alone was prescribed in 14.3% of patients.
Patient-related factors associated with prescribing noncompliance included male gender (p=0.001), higher BMI (p=0.04) and comorbidities including myocardial infarction (p=0.01), congestive heart failure (p=0.001), and metastatic tumor (p=0.01). Of note, low mobility prior to and during admission was not significantly associated with increased prescriber noncompliance.
Of the agents used for VTE prophylaxis, subcutaneous heparin was most common (65.4%), followed by enoxaparin (13.0%). Subcutaneous heparin was associated with prescribing compliance (p<0.0001) and enoxaparin was associated with noncompliance (p<0.0001). Of the full dose anticoagulants, warfarin was most common (15.1%), followed by heparin infusion (4.03%), and direct oral anticoagulants (DOACs) (4.3%). Both warfarin and heparin infusion were associated with noncompliance (p<0.0001 and p<0.0001, respectively); DOACs trended towards compliance (p=0.08).
With regards to clinical outcomes, prescribing noncompliance was associated with increased mortality (p=0.01), LOS (p<0.0001), and 30-day readmissions (Χ2=0.0004).
Conclusions: VTE prophylaxis in hospitalized older adults is a national priority. Our study found that known risk factors for VTE (mobility, BMI and comorbidities) did not improve prescriber compliance. Integrating risk assessment models for VTE prophylaxis may standardize and improve prescribing compliance.